Does metformin improve reproduction outcomes for non-obese, infertile women with polycystic ovary syndrome? Meta-analysis and systematic review

Background Polycystic ovary syndrome (PCOS) affects 10–12% of women of reproductive age. The prevalence of infertility in women with PCOS is high at between 70 and 80%. Treatment initially includes recommendations to follow preconception guidelines, such as lifestyle changes, folic acid therapy and halting the consumption of tobacco and alcohol. Management with pharmacological agents and surgical procedures have been incorporated into treatment regimens to improve fertility. Of these, metformin, an insulin sensitizer used as oral hypoglycemic agent, is gaining popularity. Objectives The aim of this study was to perform a meta-analysis of randomized controlled trials (RCTs) to evaluate the role of metformin in improving the reproduction outcomes for non-obese, infertile women with polycystic ovary syndrome. Search Methods In June 2019, we searched PubMed (from inception to present), Ovid Medline, Ovid EMBASE, Scopus, and the Cochrane library without date or language restrictions for relevant RCTs. Search was then updated in April 2020. Bibliographies of included studies were also searched for eligible studies. Selection criteria RCTs that compared the effectiveness of metformin with other modalities in treating infertility in non-obese women with PCOS were included. The eligible outcomes for inclusion were pregnancy rate, miscarriage rate, live birth rate, ovarian hyperstimulation (OHSS) and multiple pregnancy. Data Collection and analysis Data extraction and study quality assessment were performed independently by two reviewers, and any disagreements resolved by consensus or by arbitration by a third reviewer. Where two or more studies reported on the same outcome a meta-analysis was conducted using Cochrane RevMan 5. Results We found 21 RCTs which were eligible for inclusion in our systematic review, including 2638 patients with PCOS. Our meta-analysis showed that the use of metformin in non-obese women with PCOS is associated with slight increase in clinical pregnancy rate compared to placebo (47.7% vs. 42.9%) (Pooled risk ratio = 1.08 [0.82, 1.42], 95% CI, p = 0.60). It also showed that metformin is comparable to clomiphene citrate (CC) when the outcome is clinical pregnancy rate and the risk of multiple pregnancies tended to be lower (Pooled risk ratio = 0.36 [0.07, 1.92], 95% CI, p = 0.23, 3 studies). However, metformin had a higher risk of miscarriage rate (Pooled risk ratio = 2.41 [0.39, 14.86], 95% CI, p = 0.72). Furthermore, this analysis suggested that adding metformin to CC treatment decreases miscarriage risk by two folds compared to metformin alone (Pooled risk ratio = 2.67 [1.32, 5.39], 95% CI, p = 0.006) and showed no difference compared to CC alone. In comparison to letrozole, combination of metformin and CC is associated with lower clinical pregnancy rate (Pooled risk ratio = 0.52 [0.14, 1.91] 95% CI, p = 0.33) and multiple pregnancies (Pooled risk ratio = 0.45 [0.06, 3.19] 95% CI, p = 0.42). Conclusion Although this study illustrated that metformin may be better than placebo for some pregnancy outcomes, stronger, more definitive evidence from sufficiently powered trials are required before considering metformin for treating non-obese infertile women with PCOS within the current recommended guidelines.